Elsevier required licence: © <2017>. This manuscript version is made available under the 

CC‐BY‐NC‐ND 4.0 license http://creativecommons.org/licenses/by‐nc‐nd/4.0/ 

 

  

1  

Behaviour change interventions to improve medication adherence inpatients with cardiac

disease: Protocol for a mixed methods study including a pilot randomized controlled trial

Ali Hussein Alek Al-Ganmia,b,∗, Lin Perrya,c, Leila Gholizadeha,Abdulellah Modhi Alotaibia,d

a University of Technology Sydney, Faculty of Health, UTS Building 10, Level 7, 235-253 Jones Street,

Ultimo, Sydney, NSW, 2007, Australiab

b University of Baghdad, College of Nursing, Bab Al-Muadahm Square, Baghdad, Iraq

c Prince of Wales Hospital, G74, East Wing, Edmund Blacket Building, Barker Street, Randwick, NSW,

2031, Australia

d Shqra University, Faculty of Applied Health Sciences, Shqra, Saudi Arabia

  

2  

Abstract

Background: Suboptimal adherence to medication increases mortality and morbidity; individually tailored

supportive interventions can improve patients’ adherence to their medication regimens.

Aims: The study aims to use a pilot randomised controlled trial (RCT) to test the hypothesis that a theory-

based, nurse-led, multi-faceted intervention comprising motivational interviewing techniques and text

message reminders in addition to standard care will better promote medication adherence in cardiac patients

compared to standard care alone. The pilot study will assess self-reported adherence or non-adherence to

cardiovascular medication in patients referred to a cardiac rehabilitation program following hospital

admission for an acute cardiac event and test the feasibility of the intervention. The study will examine the

role of individual, behavioural and environmental factors in predicting medication non-adherence in patients

with CVD.

Methods: This is a mixed-methods study including a nested pilot RCT. Twenty-eight cardiac patients will

be recruited; an estimated sample of nine patients in each group will be required for the pilot RCT with 80%

power to detect a moderate effect size at 5% significance, and assuming 50% loss to follow-up over the six

month intervention. Participants will complete a paper-based survey (Phase one), followed by a brief semi-

structured interview (Phase two) to identify their level of adherence to medication and determine factors

predictive of non-adherence. Participants identified as ‘non-adherent’ will be eligible for the pilot

randomised trial, where they will be randomly allocated to receive either the motivational interview plus

text message reminders and standard care, or standard care alone.

Discussion: Nurse-led multi-faceted interventions have the potential to enhance adherence to cardiac

medications. The results of this study may have relevance for cardiac patients in other settings, and for long-

term medication users with other chronic diseases.

Keywords: Cardiac disease, medication adherence, pilot randomised controlled trial, motivational

interviewing, text messaging, nursing.

  

3  

Summary of Relevance

Problem

Cardiovascular disease remains one of the leading causes of morbidity and mortality in Australia and

worldwide.

What is already known

Adherence to cardiovascular medication is suboptimal, increasing the risk of mortality and morbidity.

What this paper adds

This is the first multi-method pilot randomised controlled trial of an intervention comprising

motivational interviewing techniques and text reminders delivered by nurses to promote adherence to

medication. Outcomes will support the development of a full-scale trial of the intervention in cardiac

rehabilitation and other health care areas, and to patients with other chronic conditions who take long-

term medications.

  

4  

Introduction

World-wide, the prevalence of cardiovascular disease (CVD) is increasing rapidly because of changes

in population lifestyles (Hauptman, 2008), resulting in major concerns for community health (Zhu,

Wang, Zhu, Zhou, & Wang, 2015). Cardiovascular disease has emerged as a leading cause of death and

disability in Australia (Nichols, Peterson, Alston, & Allender, 2014), affecting one in six people and

responsible for 16% of the nations’ total disease burden. It is the main reason for rehospitalisation

(Australian Institute of Health and Welfare, 2014); costs in 2004-5 of AU$5.94 billion accounted for

11% of Australia’s total health expenditure (Australian Institute of Health and Welfare, 2010). Quality

of life has been shown to improve for patients with CVD referred to a comprehensive cardiac

rehabilitation and secondary prevention program (Shepherd & While, 2012). These programs comprise

recovery and preventative activities aimed at modifying cardiac risk factors and enhancing physico-

psycho-social function to reduce the risk of subsequent cardiac events (Woodruffe et al., 2014). Cardiac

rehabilitation enables changes in patient lifestyles by providing education and development of skills to

reduce cardiovascular risk factors, and to promote adherence to prescribed medications.

Prescription medications, an important form of secondary prevention for CVD, have been a key

factor in the 20% reduction seen in mortality rates within one year of diagnosis of acute myocardial

infarction (AMI) (Chase, Bogener, Ruppar, & Conn, 2016); and the 28% reduction within three months

of AMI between 2006 and 2007 (Kolandaivelu, Leiden, Gara, & Bhatt, 2014). While cardiac

medications have been shown to be effective for symptom management and slowing the progression of

CVD, consistent adherence to prescribed medication regimes is required to achieve these effects (Hunt

et al., 2009).

The World Health Organisation (WHO) defines adherence as ‘the extent to which a person’s

behaviour (taking medications, following a recommended diet and/or executing lifestyle changes)

corresponds with the agreed recommendations of a healthcare provider’ (Sabaté, 2003, p. 3). Adherence

to medications is a challenge, particularly for patients with cardiac disease who often require multiple

medications for prolonged periods (World Health Organisation, 2003). The risk of mortality and

morbidity in such patients increases if adherence to prescribed medication is suboptimal (Hope, Wu,

  

5  

Tu, Young, & Murray, 2004); yet reported rates of non-adherence vary from 33% to more than 50%

(Li, Kuo, Hwang, & Hsu, 2012; Munger, Van Tassell, & LaFleur, 2007; Shah, Desai, Gajjar, & Shah,

2013), contributing to increased numbers of CVD-related Emergency Department visits,

hospitalisation, reduced health and well-being, augmented healthcare costs and risk of death (Mukhtar,

Weinman, & Jackson, 2014; Whittle et al., 2016). It is therefore important to identify the factors that

influence medication adherence (Munger et al., 2007) and provide tailored interventions to improve

patients’ medication-taking behaviours (Santo et al., 2016).

Background

Medication adherence is linked to better clinical outcomes among patients with heart disease, reducing

the risk of hospital readmission and death (Ruppar, Cooper, Mehr, Delgado, & Dunbar Jacob, 2016).

Suboptimal medication adherence is a multidimensional issue. Socio-economic and patient-related

factors include low educational levels, inadequate knowledge about disease and medications, beliefs

about medications and patients’ motivation to manage their illness and improve their overall health

(Broekmans, Dobbels, Milisen, Morlion, & Vanderschueren, 2010). Lack of social support and

psychological, cognitive or medical vulnerability can also play a part (Kronish & Ye, 2013). Factors

shown to predict medication non-adherence include low self-efficacy, attitudes and beliefs about

medications, low perceived behavioural control, and lack of social support (Morrison et al., 2015).

Patients with concerns about their medications are more likely to report forgetting to take them or

intentionally skipping doses (World Health Organisation, 2003). Older people and people in poor health

or with co-morbidities are less likely to successfully self-administer (Krueger et al., 2015). Medication

self-efficacy and beliefs about medications may also influence the adoption and maintenance of

medication adherence behaviours (Bane, Hughe, & McElnay, 2006), and these factors can be affected

by psycho-social factors such as the perceived level of social support and mood (Cha, Erlen, Kim,

Sereika, & Caruthers, 2008).

Exploring patients’ emotions and beliefs while helping to strengthen their self-efficacy may

minimise barriers to behavioural change and motivate them to adhere to their medications (Riegel,

Masterson Creber, Hill, Chittams, & Hoke, 2016). It is important to promote behavioural change by

  

6  

exploring what drives an individual patient to make changes or to maintain the status quo. This can be

achieved by applying motivational interviewing techniques that have been found to be effective in

assessing a patient’s readiness to change and subsequently moving toward change at an appropriate

time (Dart, 2010). Motivational interviewing has been shown to be effective in increasing medication

adherence in cardiac patients (Ogedegbe et al., 2008). Also, text messages have been effective in

improving the use of prescribed cardiac medications among 65% of patients at six months (Wald,

Bestwick, Raiman, Brendell, & Wald, 2014). Understanding the reasons for poor adherence may

suggest approaches to novel interventions.

Theoretical framework guiding the study

This study will use Bandura’s social cognitive theory to enhance the refinement of self-efficacy

(Bandura, 1977) and to examine the effects of individual, behavioural and environmental factors on

medication adherence. According to the theory, self-efficacy is the pivotal determinant in influencing a

person’s particular behaviour, directly affecting one’s actions and impacting on other determinants

(Bandura, 1977, 2004). Bandura (2004) notes that self-efficacy determines the expected outcomes of

people’s behaviours. Reciprocal determinism is the basic organising principle of behaviour change

proposed by this social cognitive theory, with continuous, functional interaction between the

environment, the individual and behaviour (Bandura, 1998) (Figure 1). It assumes that a change in

knowledge of health risk and benefits is essential, but requires additional impacts for change to occur

(Munro, Lewin, Swart, & Volmink, 2007). Other determinants of behaviour change include behaviours,

outcome expectations, expected benefits, beliefs and goals, and perceived facilitators and barriers. The

theory proposes that if people perceive that they have outcome control, appropriate behaviour will

follow, with sufficiently high self-confidence to overcome otherwise insurmountable barriers (Armitage

& Conner, 2000).

Using a theoretical basis to explain relationships between study variables is important when

designing effective behavioural change studies (Short, James, & Plotnikoff, 2013). Appropriately

designed interventions that employ multiple strategies, such as motivational interviewing to encourage

behaviour changes and text messaging strategies to reinforce behaviours, are likely to achieve

  

7  

significant increases in medication adherence in cardiac patients (Al Ganmi, Perry, Gholizadeh, &

Alotaibi, 2016).

Methods

Aim and hypothesis

The study aims to use a pilot randomised controlled trial (RCT) to test the hypothesis that a theory

based, nurse-led, multi-faceted intervention comprising motivational interviewing techniques and text

message reminders in addition to standard care will better promote medication adherence in cardiac

patients compared to standard care alone. Underpinning evidence and assumptions of the hypothesis

are that:

1) A high proportion of patients fail to adhere to their cardiovascular medication regimens.

2) Medication adherence self-efficacy, patient beliefs, and lack of social support are predictive factors

associated with non- adherence to medication.

3) High quality evidence supports the use of multi-faceted interventions comprising of motivational

interviewing counselling combined with text messaging reminders to promote medication adherence.

The pilot study will assess self-reported adherence or non-adherence to cardiovascular medication in

patients referred to a cardiac rehabilitation program following hospital admission for an acute cardiac

event and test the feasibility of the intervention. Medication non-adherence has been defined as ‘taking

less than 80% of prescribed doses and can also include taking too many doses’ and it is associated with

an increased risk of poor health, adverse clinical events and death (Nieuwlaat et al., 2014). The study

will examine the role of individual, behavioural and environmental factors in predicting medication

non-adherence in patients with CVD.

Study design

This is a mixed methods study which includes a nested, pilot RCT. Both qualitative and quantitative

methods are required to address the study’s aims and hypothesis. The use of mixed methods is valuable

to provide a fuller picture of the topic and to transcend the limitations of each of the methods used

singly (Borkan, 2004; Creswell, Fetters, & Ivankova, 2004).

  

8  

In this study, data collection will occur sequentially: quantitative data collection followed by

qualitative data collection will occur for the exploratory phases that will provide the explanatory power

in support of the final (RCT) intervention testing phase: (QUAN + qual) → QUAN. Results from the

exploratory phases (survey design and semi-structured interviews) will inform the fine tuning of the

intervention in the third phase.

This multi-method study will entail three interrelated phases:

Phase one: The survey is designed to identify cardiac patients’ patterns of medication

adherence and their associated degree of adherence. The purpose of the survey is to gather quantitative

data about medication adherence, patient behaviours, beliefs and other associated factors, including

demographic data which will inform intervention and form baseline data.

Phase two: A descriptive qualitative study will explore the phenomena of cardiac patients’

adherence to medications and how they respond to factors that affect medication adherence. Using semi-

structured interviews, patients’ views of their cardiac medications will be explored as well as the factors

that influence their medication adherence.

A semi-structured interview format was chosen because of its flexibility in collecting self-reporting

data (Cohen & Crabtree, 2006), enabling participants to talk freely about issues related to their

medication adherence and telling stories in their own words. Semi-structured in-depth interviews will

involve open-ended questions to elicit detailed narratives and stories (Whiting, 2008). This will provide

in-depth understanding to supplement and explain quantitative results from Phase one.

Details of Phase one and Phase two data will be used to inform motivational interviewing

techniques and to tailor the interventions to individual patients’ needs to better support their adherence

to cardiac medications.

Phase three, a pilot RCT: Participants identified as non-adherent to their medications in Phase

one (based on Medication Adherence Questionnaire [MAQ] scores) and Phase two data (themes and

patterns of medication adherence/non-adherence) will be invited to take part in this pilot trial. This will

pilot test the feasibility and effectiveness of a multi-faceted intervention strategy, including

motivational interviews plus text message reminders, to influence adherence to their cardiac medication

regimes. The RCT will test and compare differences in outcomes between the standard care delivered

  

9  

by a cardiac rehabilitation program and the same program augmented by this multi-faceted intervention.

The feasibility of conducting a multi-faceted behavioural intervention in a busy cardiac rehabilitation

setting will be investigated. We also sought data to determine the effect size for sample-size calculation

as we could only extrapolate an estimated effect size from a somewhat similar trial for the sample size

calculation of this pilot study. Data are currently lacking for full-scale trial of this approach among

cardiac rehabilitation patients.

Study setting and participants

The study will be carried out in the cardiac rehabilitation centre of a tertiary referral hospital in Sydney,

Australia. Participants will be cardiac patients referred to a cardiac rehabilitation program following

their admission for an acute cardiac event; those participating in the pilot trial will be randomly allocated

to receive either the intervention plus standard care or standard care only. Inclusion criteria are: 18 years

of age or older; diagnosed with cardiac disease and referred to the hospital cardiac rehabilitation

program; currently taking at least one cardio-protective medication, and having primary responsibility

for taking their own medications (i.e. not reliant on a carer). Participants must be able to read, speak

and understand English, have a personal mobile phone, and be able to receive and reply to phone calls

and text messages. Patients who are blind, deaf or unable to consent to receiving text messages will be

excluded. Those clinically judged to have cognitive impairment that limits their ability to understand

and answer the study questions will also be excluded.

Sample size determination

The sample size for the pilot RCT was based on data from Ma, Zhou, Zhou, and Huang (2014). Using

a two-sided test, moderate effect size with α = 0.05, and power = 0.80, nine participants are required in

each group. Allowing for 50% loss, a total of 28 cardiac patients will be needed for the pilot RCT.

Around 350 patients per year (30 per month) are estimated to attend the cardiac rehabilitation centre.

Pilot data indicate around one third are considered non-adherent to medications. If 50% of eligible

patients agree to participate, recruitment will progress at five participants per month (a rate of

recruitment easily manageable for the researcher): it will take six months to enrol 28 participants, so

  

10  

the study will take approximately six months for recruitment and six more months for intervention and

follow-up.

Study procedures and data collection

Data collection will take place at baseline and at six months, and changes in medication adherence will

be compared within and between groups to determine any intervention effect. The Consolidated

Standard of Reporting Trials (CONSORT) diagram of study processes is presented in Figure 2. Baseline

survey and interview data will be collected during patients’ hospital attendance for a cardiac

rehabilitation session. Pilot RCT participants will repeat the Phase one survey via telephone interview

at six months post-randomisation to collect the follow-up data; three additional open-ended questions

(see Appendix 1) will explore participants’ experiences of the motivational interview and the text

messaging, and enquire whether anything else might have helped adherence.

In Phase two each interview will involve a face-to-face meeting for an interview guided by a

set of questions or topics (Polit & Beck, 2004). Interviews will be arranged with participants using

preferred communication methods with reminders two days beforehand. The guide for the individual

semi-structured interviews can be found in Appendix 2. Data collection, transcription and analysis will

be conducted, with questions modified in light of responses and emergent themes (Braun & Clarke,

2006).

Recruitment, enrolment and consent

Recruitment will occur under the supervision of the clinical nurse consultant for cardiac rehabilitation,

with the agreement of the director of nursing and the cardiology consultant. Patients referred to the

cardiac rehabilitation program will be screened using the study inclusion/ exclusion criteria by the

clinical nurse consultant at their first session. Eligible members of a consecutive cohort of patients who

express interest in the project will be referred to the researcher, who will explain the study phases and

provide an information statement and consent form. Completion of the survey will identify patients who

are non-adherent with at least one cardiac medication, based on their answers to the MAQ with their

non-adherence confirmed and detailed at Phase two by semi-structured interview. At the conclusion of

  

11  

the semi-structured interview (Phase two), eligible participants will be informed about Phase three, the

pilot RCT, and invited to participate.

The researcher will obtain written informed consent from participants for Phases one and two;

similar consent to participate in the pilot trial will be obtained separately. Participants will receive a

folder at enrolment, including written information about the study and a copy of their consent form, to

supplement the face-to-face explanation of the study.

Randomisation and blinding processes

In the pilot RCT, participants will be randomly assigned to either intervention or control group after

they have consented to the trial phase. A computerised random number generator will be used to

generate the random sequence. Using permutated blocks equal numbers of participants will be assigned

to each group. Sequentially numbered sealed envelopes will be used to conceal the sequence until

participants are assigned. Neither the participants nor the researcher can be blinded to the group

allocation because of the nature of the behavioural intervention. A behavioural intervention such as

motivational interviewing is not easily delivered blinded, and it is unlikely that cardiac patients in a

cardiac rehabilitation program would not know which intervention they are receiving (Page & Persch,

2013). Outcome data are self-reported, but the assessor who will collect the outcome data will be

blinded to study allocation to reduce potential bias.

The study intervention

Approaches such as motivational interviewing aim to encourage behaviour change through influencing

individual’s feelings, thoughts, and confidence to change their behaviours, for example: adhere to a

recommended medical regimen. The technique facilitates behaviour change by resolving patients’

ambivalence to change (Miller & Rollnick, 2002). Motivational interviewing has been incorporated into

several successful medication adherence interventions (Solomon et al., 2009). Using this technique, a

wide variety of medication non-adherence factors are targeted, enabling patients to reflect on perceived

barriers and search for solutions (Easthall, Song, & Bhattacharya, 2013). The interviewing ‘elicits a

range of possible actions and affirms the patient’s autonomy to make informed choices’ (Riegel et al.,

2016, p. 3). Self-efficacy, one of the targeted factors, predicts adherence to medications, and this may

  

12  

be mediated by patients’ beliefs about their ability to cope, to locate optimistic changes in their social

relations and to improve their individual confidence (Luszczynska, Sarkar, & Knoll, 2007). Social

cognitive theory has previously been applied to examine medication adherence among CVD patients

(Haskell et al., 1994), to foster individuals’ medication adherence by setting goals and creating firm

commitments to them, thereby encouraging patients to exercise more control over behaviours. In turn

patients can come to believe they have a higher ability to adhere to their medications (Smith, Rublein,

Marcus, Brock, & Chesney, 2003).

Text messaging has been widely used to improve health behaviours across multiple diseases

(Spoelstra et al., 2015). It has been used to improve medication adherence by affecting different

elements of cognition and evoking attention and an automatic response (Yiend, 2010). Text Messaging

is widely available low-cost, patient friendly, requires low technological expertise and is applicable in

a diverse range of health behaviours, acting as a safety-net for medication reminders (Cole-Lewis &

Kershaw, 2010; Zallman, Bearse, West, Bor, & McCormick, 2016).

Intervention arm

Participants in the intervention group will receive current standard care through the cardiac

rehabilitation program plus behavioural counselling about medication adherence using motivational

interviewing and text reminders. The current cardiac rehabilitation program includes a single one-hour

information session on cardiac medication. The content received by both group participants will be

recorded and compared to check for equivalence between the groups.

Each patient in the intervention group will receive approximately 30 to 40 minutes of a single

motivational interview-style counselling session (Ma et al., 2014). The essence of motivational

interviewing is for the counsellor to be simultaneously sympathetic and supportive, as well as directive

in moving patients toward behaviour change by strengthening their reasons to change (Levensky,

Forcehimes, O'Donohue, & Beitz, 2007). As part of the counselling, the researcher will provide

information that the patient may need, and explore barriers that keep the patient from adhering to the

medication regimen (Dart, 2010). A structured counselling script will be designed that can be tailored

to medication adherence characteristics of individual patients (Dart, 2010). Examples of the content of

  

13  

the session are available in Appendix 3. The sessions will be audiotaped and reviewed by a clinician

qualified in motivational interviewing to ensure fidelity (Ma et al., 2014).

Text message reminders will be sent daily for the first two weeks, then on alternate days for

fortnight, then once a week for the next five months, for a total of six months (Wald et al., 2014). The

content will vary according to each patient’s non-adherence factors. Examples are supplied in Table 1.

Control arm

For the period of the study, patients in the control arm will be provided only with current standard care,

which includes the provision of cardiac care knowledge and recommendations to enhance medication

adherence and to promote healthy lifestyles; they will complete the same surveys as the intervention

arm at baseline and at six months.

Outcomes

The primary outcome is medication adherence/ non-adherence at six months. Medication adherence

will be determined based on participants’ responses to the MAQ (Morisky, Green, & Levine, 1986).

Patients with a sum score of 1-2 will be considered ‘adherent’; those with a score of 3 - 4 as ‘non-

adherent’ (Morisky et al., 1986).

Secondary outcomes include identification of those factors exerting a significant influence on

medication adherence, such as behaviour (self-regulation), beliefs (general harm; general overuse;

specific necessity; and specific concerns), social support and self-efficacy using a combination of

instruments, outlined below.

To assess the acceptability of the intervention at six months, the researcher will telephone each

member of the intervention group to evaluate their experience of the interview and text messages (see

Appendix 1).

Study instruments

The survey will include a set of questionnaires designed to gather data about medication adherence and

patient behaviours, beliefs and other factors associated with adherence/ non-adherence. The survey will

be paper based, self-administered, and suitable for participants with low literacy skills.

  

14  

Sociodemographic and health data will be collected: age, gender, living arrangement, level of education,

ethnicity; co-morbidities.

The use of different instruments in this study will enable comprehensive study of factors

contributing to medication non-adherence in patients with CVD, including behavioural and

psychological factors. Motivational interviews in the RCT phase will focus on the identified non-

adherence factors for each individual patient.

Medication non-adherence will be assessed by the Medication Adherence Questionnaire

(MAQ) (Morisky et al., 1986), which is designed to measure medication adherence behaviour and

barriers such as forgetfulness, carelessness, adverse effects and efficacy. It includes four simple

dichotomous questions, assigning one point for each yes response; total scores are categorised as: 0=

high; 1-2= medium, and 3-4= low medication adherence behaviours (Morisky et al., 1986).

The Adherence to Refills and Medications Scale (ARMS) (Kripalani, Risser, Gatti, & Jacobson,

2009) will be used to determine medication adherence behaviour in terms of self-regulation. The ARMS

is a 12-item scale: an eight-item medication-taking subscale assessing correct self-administration for

the prescribed medications; and a four-item prescription refill subscale evaluating the patient’s ability

to replenish medications on schedule. Each item is scored on a four-point scale ranging from l= none

of the time to 4= all the time on a four-point Likert scale, with higher numbers demonstrating better

refill ability for medications on schedule (Kripalani et al., 2009).

The Belief about Medicine Questionnaire (BaMQ) (Horne, Weinman, & Hankins, 1999) elicits

information on patients’ beliefs about medications which may be adherence-related. It identifies

whether the patient believes in the necessity of their medicines or has concerns about them. The study

will use the short (eight-item) version of the BaMQ developed by Horne et al. (1999), composed of four

subscales of two items each, assessing: specific necessity, specific concerns, general overuse and

general harm. Answers range from l= strongly disagree to 5= strongly agree on a five-point Likert scale;

scores are summed to derive a total score, with higher scores indicating more positive beliefs.

The Medication Adherence Self-Efficacy Scale-Revised (MASESE-R) (Fernandez, Chaplin,

Schoenthaler, & Ogedegbe, 2008) consists of 13 items that evaluate an individual’s ability to adhere to

medications under various challenging circumstances. Twelve items examine patients’ confidence in

  

15  

taking medications in specific circumstances (e.g. with family, in public places, feeling well), and one

assess the ability to take medications as part of the everyday routine. Each item is scored on a four-

point Likert scale, ranging from 0= not at all sure to 3= extremely sure. A single score is derived from

the mean of all items, with greater self-efficacy indicated by high scores (Fernandez et al., 2008).

The Medication Specific Social Support (MSSS) scale (Lehavot et al., 2011) is an eight-item

survey of medication-specific social support to identify how often others help patients with their

medication, scored for each item ranging from 0= never to 4= very often. A single mean score is

presented as a medication-specific index of support (Lehavot et al., 2011).

Validity and reliability

The MAQ is a validated four-item tool suitable for a wide range of conditions involving cardiac

diseases (Nguyen, Caze, & Cottrell, 2014). Its validity and reliability has been determined in patients

with hypertension, with a reported acceptable internal consistency of α = 0.61, sensitivity 0.81 and

specificity 0.44 (Lavsa, Holzworth, & Ansani, 2011); it has also been validated in patients with heart

failure, CVD and dyslipidemia (Afonso, Nassif, Aranha, DeLor, & Cardozo, 2006). MAQ score was

found to be a significant independent predictor of cardiovascular nonadherence in a multivariate logistic

regression model (Shalansky, Levy, & Ignaszewski, 2004) (Table 2).

The ARMS subscales are highly correlated with the Morisky medication adherence four-item

scale, and with medication refill adherence (Kripalani et al., 2009). The ARMS has been validated in

patients with cardiovascular disease and other chronic diseases (Kripalani, Henderson, Jacobson, &

Vaccarino, 2008). Among patients with low literacy skills (Kripalani et al., 2009) it has demonstrated

high internal consistency using Cronbach’s α and test–retest reliability, set out in Table 2 (Kripalani et

al., 2009).

The BaMQ has been shown to correlate significantly with other adherence-related

questionnaires such as MAQ, the Morisky medication adherence scale, and the medication adherence

rating scale (MARS-5) (Gatti, Jacobson, Gazmararian, Schmotzer, & Kripalani, 2009; Horne et al.,

1999; Mårdby, Åkerlind, & Jörgensen, 2007). Each BMQ subscale has been evaluated for internal

consistency using Cronbach's α (Horne et al., 1999). Demonstrating criterion-related validity, the four

  

16  

BaMQ categories correlated highly with patients’ beliefs about the adverse effects of medication and

specific- concerns as assessed by the Sensitive-Soma Scale administered to general medical and cardiac

groups (Table 2) (Horne et al., 1999).

The MASES-R has been found to correlate significantly with electronic medication adherence

records (MEMS) at three-months, confirming its predictive validity (Fernandez et al., 2008). The

concurrent validity of the MASES-R has also been confirmed (Table 2).

Data Analysis

Phase one: Descriptive statistics will be used to analyse data related to the patients’ baseline

characteristics. Data will be checked and cleaned prior to entry into SPSS for Windows version 23.

Measures of central tendency and dispersion will describe the values of medication adherence,

medication adherence self-efficacy, and beliefs about medication. Bivariate analyses will be conducted

to examine factors potentially associated with medication non-adherence. Factors and behaviours

related to medication non-adherence will be examined by logistic regression. Two sided testes will be

conducted with significance set at .05.

Phase two: Qualitative data will be analysed inductively using thematic analysis (Sim, 1998). This

approach will focus on recognising, analysing and reporting recurrent patterns (themes) and

subcategories within the data (Liamputtong, 2013). Simultaneous collection, transcription and analysis

allows the researcher to build on emerging themes (Polit & Beck, 2014). Data will be examined and

compared to identify similarities and differences by reading and searching within transcripts and across

the data set (Polit & Beck, 2014). QSR NVivo software will be used to code and construct thematic

analysis (Braun & Clarke, 2006). A similar but separate process will be employed with the data

collected in response to the open questions at six months.

Pilot RCT phase: Data will be analysed according to the intention to treat principles. Using survey and

demographic data, the two groups will be compared at baseline and any differences taken into account

during the outcome assessment. Paired samples t-test analysis will be used to test within-group

differences on the medication adherence questionnaire scores (MSSS, MAQ, ARMS, BaMQ, and

MASER-R), and independent sample t-test will compare the scores of the intervention and control

  

17  

groups. Multinomial regression analysis will be applied to identify variables that significantly influence

adherence to medication, such as self-efficacy, beliefs, level of confidence, or social support. Variables

exerting significant influence on medication adherence in bivariate analysis will be entered in the

regression analysis, with significance set at P<0.25 for the preliminary bivariate analysis and P <0.05

for the regression analysis (Polit., 1996).

Ethical considerations

Approvals to conduct this study were granted by the appropriate health district and university Human

Research Ethics Committees in June 2016 (reference numbers: 16/085 (HREC/16/POWH/218; ETH16-

0635). The study is registered as a clinical trial (ACTRN12616000910404) on www.anzctr.org.au.

Discussion

Patients with CVD often have multiple chronic illnesses requiring multiple prescriptions. Studies of

long-term medication adherence outcomes are limited (Bansilal et al., 2016) although non-adherence is

common in these patients and accounts for substantial morbidity and mortality (Albert, 2008). Adequate

medication adherence may help improve quality of life by improving disease outcomes (Luszczynska

et al., 2007). Better medication adherence may be achieved with interventions that address the known

multiple factors of non-adherence, including lack of patient knowledge of perceived benefits, the

perceived harm of medications, poor medication management and inadequate social support (Calvert et

al., 2012).

Medication adherence interventions for patients with CVD have been shown to be effective

when delivered by nurses, who should play an active role in designing and applying such interventions

(Albert, 2008; Chase et al., 2016). A recent systematic review emphasised that cardiac rehabilitation

and prevention programs should encompass a dedicated strategy for medication adherence; at present,

the majority of these programs do not measure and report adherence outcomes appropriately (Santo et

al., 2016). There is scope for medication adherence interventions using techniques focused on

promoting behavioural change to become part of routine health care (Easthall et al., 2013). This study

aims to evaluate the effectiveness of theory-based, nurse-led, multi-faceted interventions for medication

  

18  

adherence in a robust pilot trial using a RCT design, in line with the findings and recommendations of

the recent review of medication adherence interventions for this population by Al Ganmi et al. (2016).

Limitations

This study has some potential limitations. Firstly, use of a multi-faceted intervention, while

recommended as likely to increase the success of intervention as a whole, challenges researchers to

identify the contribution of the individual components. This study attempts to address this by seeking

participants’ experiences of each separate component of the intervention. Secondly, measuring

medication adherence through self-reporting may introduce a response bias presenting potential

problems with reliability. Also, social desirability bias is possible due to lack of blinding. Self-reporting

is a commonly used method of outcome assessment for medication adherence as it is acceptable to

participants and imposes a minimal burden. It also has the potential to explore medication adherence

behaviour, adherence-related barriers and beliefs about medicines which may support beneficial

medication adherence assessment in chronic disease management. Finally, participants are

predominantly older adults who will be asked to spend 20 - 30 minutes answering 45 questions in Phase

one (repeated at six months for participants in the intervention group) and 30-45 minutes in the Phase

two interview. This may challenge their stamina, patience and tolerance.

Conclusion

Medication non-adherence among patients with CVD is a major problem. Multi-faceted medication

adherence interventions comprising motivational interviews and text reminders may improve adherence

to cardiac medication regimens by targeting individual behaviour change. This pilot study will provide

important information about techniques appropriate for use by nurses to support medication compliance

in their patients. Findings may support development of further trials of this intervention in out-patient

cardiac care settings ahead of translation into routine clinical care. It may also be suitable for

implementing in other health care areas and long-term patient groups.

If this and further trials are successful, important next steps will be to consider the potential for

widespread roll-out, including training nurses in motivational interviewing techniques. Software

compatible with organisations’ routine data programs will be required for automatic delivery of text

  

19  

messages. The time requirement of the intervention will need to be considered for staffing rosters and

patient appointments. A full cost-effectiveness analysis is therefore recommended.

This protocol sets out the essential first steps for what may be an exciting new development in

the contribution of nursing staff to delivering a substantial benefit to a sizeable patient group where

there is clear potential for significant improvement in adherence.

Acknowledgments

The authors declare that there are no conflicts of interest that are directly relevant to the content of this

review. This research was enabled by a doctoral scholarship for the first author from the Iraqi Cultural

Attaché/ Canberra, Australia.

  

20  

References

Afonso,  N.  M.,  Nassif,  G.,  Aranha,  A.,  DeLor,  B.,  &  Cardozo,  L.  J.  (2006).  Low‐density  lipoprotein cholesterol goal attainment among high‐risk patients: does a combined intervention targeting patients and providers work. Am J Manag Care, 12(10), 589‐594.  

Al‐Ganmi,  A.  H.,  Perry,  L.,  Gholizadeh,  L.,  &  Alotaibi,  A.  M.  (2016).  Cardiovascular  medication adherence among patients with  cardiac disease:  a  systematic  review.  Journal of Advanced Nursing, 72(12), 3001‐3014.  

Albert, N. M. (2008). Improving Medication Adherence in Chronic Cardiovascular Disease. Critical Care Nurse, 28(5), 54‐65.  

Armitage,  C.  J., &  Conner, M.  (2000).  Social  cognition models  and  health  behaviour:  A  structured review. Psychol Health, 15. doi:10.1080/08870440008400299 

Australian  Institute  of  Health  and  Welfare.  (2010).  Australia’s  health  2010.      Retrieved  from www.aihw.gov.au/publication‐detail/?id=644246837 

Australian  Institute of Health and Welfare.  (2014). Leading causes of death in Australia. Australia’s health  series  no.  14.  Cat.  no.  AUS  178.    Retrieved  from www.aihw.gov.au/workarea/downloadasset.aspx?id=60129548150 

Bandura, A. (1977). Self‐efficacy: toward a unifying theory of behavioral change. Psychological review, 84(2), 191.  

Bandura, A. (1998). Health promotion from the perspective of social cognitive theory. Psychology & Health, 13(4), 623‐649. doi:10.1080/08870449808407422 

Bandura, A. (2004). Health Promotion by Social Cognitive Means. Health Education & Behavior, 31(2), 143‐164. doi:10.1177/1090198104263660 

Bane, C., Hughe, C. M., & McElnay, J. C. (2006). Determinants of medication adherence in hypertensive patients: an application of self‐efficacy and the Theory of Planned Behaviour.  International Journal of Pharmacy Practice, 14(3), 197‐204. doi:10.1211/ijpp.14.3.0006 

Bansilal, S., Castellano, J. M., Garrido, E., Wei, H. G., Freeman, A., Spettell, C., . . . Fuster, V. (2016). Assessing  the  Impact  of  Medication Adherence  on  Long‐Term  Cardiovascular  Outcomes. Journal  of  the  American  College  of  Cardiology,  68(8),  789‐801. doi:http://dx.doi.org/10.1016/j.jacc.2016.06.005 

Borkan, J. M. (2004). Mixed methods studies: a foundation for primary care research. Ann Fam Med, 2(1), 4‐6.  

Braun,  V.,  &  Clarke,  V.  (2006).  Using  thematic  analysis  in  psychology.  Qualitative  Research  in Psychology, 3(2), 77‐101. doi:10.1191/1478088706qp063oa 

Broekmans, S., Dobbels,  F., Milisen, K., Morlion, B., & Vanderschueren, S.  (2010). Determinants of medication underuse and medication overuse in patients with chronic non‐malignant pain: A multicenter  study.  International  Journal  of  Nursing  Studies,  47(11),  1408‐1417. doi:http://dx.doi.org/10.1016/j.ijnurstu.2010.03.014 

Calvert, S. B., Kramer, J. M., Anstrom, K. J., Kaltenbach, L. A., Stafford, J. A., & Allen LaPointe, N. M. (2012).  Patient‐focused  intervention  to  improve  long‐term  adherence  to  evidence‐based medications:  A  randomized  trial.  American  Heart  Journal,  163(4),  657‐665.e651. doi:http://dx.doi.org/10.1016/j.ahj.2012.01.019 

Cha, E., Erlen, J. A., Kim, K. H., Sereika, S. M., & Caruthers, D. (2008). Mediating roles of medication –taking  self‐efficacy  and  depressive  symptoms  on  self‐reported  medication  adherence  in persons with  HIV:  A  questionnaire  survey.  International  Journal  of  Nursing  Studies,  45(8), 1175‐1184. doi:10.1016/j.ijnurstu.2007.08.003 

Chase, J., Bogener, J. L., Ruppar, T. M., & Conn, V. S. (2016). The Effectiveness of Medication Adherence Interventions Among Patients With Coronary Artery Disease: A Meta‐analysis. The Journal of cardiovascular nursing, 31(4), 357‐366.  

Cohen, D., & Crabtree, B. (2006). Qualitative research guidelines project. 

  

21  

Cole‐Lewis,  H.,  &  Kershaw,  T.  (2010).  Text  Messaging  as  a  Tool  for  Behavior  Change  in  Disease Prevention  and  Management.  Epidemiologic  Reviews,  32(1),  56‐69. doi:10.1093/epirev/mxq004 

Creswell, J. W., Fetters, M. D., & Ivankova, N. V. (2004). Designing a Mixed Methods Study in Primary Care. Ann Fam Med, 2(1), 7‐12.  

Dart, M. A.  (2010). Motivational  interviewing in nursing practice: Empowering the patient:  Jones & Bartlett Publishers. 

Easthall, C., Song, F., & Bhattacharya, D. (2013). A meta‐analysis of cognitive‐based behaviour change techniques  as  interventions  to  improve  medication  adherence.  BMJ  open,  3(8). doi:10.1136/bmjopen‐2013‐002749 

Fernandez, S., Chaplin, W., Schoenthaler, A., & Ogedegbe, G. (2008). Revision and validation of the medication adherence self‐efficacy scale (MASES) in hypertensive African Americans. Journal of Behavioral Medicine, 31(6), 453‐462. doi:10.1007/s10865‐008‐9170‐7 

Gatti, M. E.,  Jacobson, K. L., Gazmararian, J. A., Schmotzer, B., & Kripalani, S.  (2009). Relationships between  beliefs  about  medications  and  adherence.  American  Journal  Of  Health‐System Pharmacy:  AJHP:  Official  Journal  Of  The  American  Society  Of  Health‐System  Pharmacists, 66(7), 657‐664. doi:10.2146/ajhp080064 

Haskell, W. L., Alderman, E. L., Fair, J. M., Maron, D. J., Mackey, S. F., Superko, H. R., . . . Krauss, R. M. (1994).  Effects  of  intensive multiple  risk  factor  reduction  on  coronary  atherosclerosis  and clinical cardiac events in men and women with coronary artery disease. The Stanford Coronary Risk Intervention Project (SCRIP). Circulation, 89(3), 975‐990. doi:10.1161/01.cir.89.3.975 

Hauptman, P.  (2008). Medication adherence  in heart  failure. Heart Failure Reviews, 13(1), 99‐106. doi:10.1007/s10741‐007‐9020‐7 

Hope, C. J., Wu, J., Tu, W., Young, J., & Murray, M. D. (2004). Association of medication adherence, knowledge,  and  skills  with  emergency  department  visits  by  adults  50  years  or  older with congestive heart failure. American Journal Of Health‐System Pharmacy: AJHP: Official Journal Of The American Society Of Health‐System Pharmacists, 61(19), 2043‐2049.  

Horne, R., Weinman, J., & Hankins, M. (1999). THE BELIEFS ABOUT MEDICINES QUESTIONNAIRE: THE DEVELOPMENT  AND  EVALUATION  OF  A  NEW  METHOD  FOR  ASSESSING  THE  COGNITIVE REPRESENTATION OF MEDICATION. Psychology & Health, 14(1), 1.  

Hunt, S. A., Abraham, W. T., Chin, M. H., Feldman, A. M., Francis, G. S., Ganiats, T. G., . . . Yancy, C. W. (2009).  2009  Focused  Update  Incorporated  Into  the  ACC/AHA  2005  Guidelines  for  the Diagnosis and Management of Heart Failure in Adults: A Report of the American College of Cardiology  Foundation/American  Heart  Association  Task  Force  on  Practice  Guidelines Developed in Collaboration With the International Society for Heart and Lung Transplantation. Journal  of  the  American  College  of  Cardiology,  53(15),  e1‐e90. doi:https://doi.org/10.1016/j.jacc.2008.11.013 

Kolandaivelu, K.,  Leiden, B. B., Gara, P.  T., & Bhatt, D.  L.  (2014). Non‐adherence  to  cardiovascular medications. European Heart Journal, 35(46), 3267‐3276.  

Kripalani, S., Henderson, L. E., Jacobson, T. A., & Vaccarino, V. (2008). Medication use among inner‐city patients after hospital discharge: patient‐reported barriers and solutions. Paper presented at the Mayo Clinic Proceedings. 

Kripalani,  S.,  Risser,  J.,  Gatti, M.  E., &  Jacobson,  T.  A.  (2009).  Development  and  evaluation  of  the Adherence to Refills and Medications Scale (ARMS) among low‐literacy patients with chronic disease. Value In Health: The Journal Of The International Society For Pharmacoeconomics And Outcomes Research, 12(1), 118‐123. doi:10.1111/j.1524‐4733.2008.00400.x 

Kronish, I. M., & Ye, S. (2013). Adherence to cardiovascular medications: lessons learned and future directions.  Progress  in  Cardiovascular  Diseases,  55(6),  590‐600. doi:10.1016/j.pcad.2013.02.001 

  

22  

Krueger, K., Botermann, L., Schorr, S. G., Griese‐Mammen, N., Laufs, U., & Schulz, M.  (2015). Age‐related medication adherence in patients with chronic heart failure: A systematic literature review. International Journal of Cardiology, 184, 728‐735. doi:10.1016/j.ijcard.2015.03.042 

Lavsa,  S.  M.,  Holzworth,  A.,  &  Ansani,  N.  T.  (2011).  Selection  of  a  validated  scale  for  measuring medication adherence. Journal Of The American Pharmacists Association: Japha, 51(1), 90‐94. doi:10.1331/JAPhA.2011.09154 

Lehavot, K., Huh, D., Walters, K. L., King, K. M., Andrasik, M. P., & Simoni, J. M. (2011). Buffering Effects of General and Medication‐Specific Social Support on the Association Between Substance Use and  HIV  Medication  Adherence.  AIDS  Patient  Care  And  Stds,  25(3),  181‐189. doi:10.1089/apc.2010.0314 

Levensky, E. R., Forcehimes, A., O'Donohue, W. T., & Beitz, K. (2007). Motivational Interviewing: An evidence‐based approach to counseling helps patients follow treatment recommendations. AJN The American Journal of Nursing, 107(10), 50‐58.  

Li, W. W., Kuo, C. T., Hwang, S. L., & Hsu, H. T. (2012). Factors related to medication non‐adherence for patients with hypertension in Taiwan. Journal of Clinical Nursing, 21(13‐14), 1816‐1824. doi:10.1111/j.1365‐2702.2012.04088.x 

Liamputtong, P. (2013). Research methods  in health :  foundations for evidence‐based practice  (2nd edition ed.). South Melbourne, Victoria: Oxford University Press. 

Luszczynska,  A.,  Sarkar,  Y.,  &  Knoll,  N.  (2007).  Received  social  support,  self‐efficacy,  and  finding benefits  in  disease  as  predictors  of  physical  functioning  and  adherence  to  antiretroviral therapy.  Patient  Education  and  Counseling,  66(1),  37‐42. doi:http://dx.doi.org/10.1016/j.pec.2006.10.002 

Ma, C., Zhou, Y., Zhou, W., & Huang, C. (2014). Evaluation of the effect of motivational interviewing counselling  on  hypertension  care.  Patient  Education  and  Counseling,  95(2),  231‐237. doi:http://dx.doi.org/10.1016/j.pec.2014.01.011 

Mårdby,  A.‐C.,  Åkerlind,  I.,  &  Jörgensen,  T.  (2007).  Beliefs  about  medicines  and  self‐reported adherence  among  pharmacy  clients.  Patient  Education  and  Counseling,  69(1–3),  158‐164. doi:http://dx.doi.org/10.1016/j.pec.2007.08.011 

Miller, W. R., & Rollnick, S. (2002). Motivational interviewing: preparing people for change. New York: Guilford Press.  

Morisky,  D.  E.,  Green,  L. W., &  Levine,  D. M.  (1986).  Concurrent  and  predictive  validity  of  a  self‐reported measure of medication adherence. Med Care, 24(1), 67‐74.  

Morrison, V. L., Holmes, E. A. F., Parveen, S., Plumpton, C. O., Clyne, W., De Geest, S., . . . Hughes, D. A.  (2015).  Predictors  of  Self‐Reported  Adherence  to  Antihypertensive  Medicines:  A Multinational,  Cross‐Sectional  Survey.  Value  in  Health,  18(2),  206‐216. doi:http://dx.doi.org/10.1016/j.jval.2014.12.013 

Mukhtar, O., Weinman, J., & Jackson, S. (2014). Intentional Non‐Adherence to Medications by Older Adults. Drugs & Aging, 31(3), 149‐157. doi:10.1007/s40266‐014‐0153‐9 

Munger, M. A., Van Tassell, B. W., & LaFleur, J. (2007). Medication Nonadherence: An Unrecognized Cardiovascular Risk Factor. Medscape general medicine, 9(3), 58‐58.  

Munro, S., Lewin, S., Swart, T., & Volmink, J. (2007). A review of health behaviour theories: how useful are these for developing  interventions to promote  long‐term medication adherence for TB and HIV/AIDS? BMC Public Health, 7(1), 1‐16. doi:10.1186/1471‐2458‐7‐104 

Nguyen, T. M. U., Caze, A. L., & Cottrell, N. (2014). What are validated self‐report adherence scales really measuring?: a systematic review. British Journal Of Clinical Pharmacology, 77(3), 427‐445.  

Nichols, M., Peterson, K., Alston, L., & Allender, S.  (2014). Australian heart disease statistics 2014.   Retrieved  from https://heartfoundation.org.au/images/uploads/publications/HeartStats_2014_web.pdf 

  

23  

Nieuwlaat,  Wilczynski  N,  Navarro  T,  Hobson  N,  Jeffery  R,  Keepanasseril  A,  .  .  .  B,  H.  R.  (2014). Interventions  for  enhancing  medication  adherence.  Cochrane  Database  of  Systematic Reviews(11). doi:10.1002/14651858.CD000011.pub4 

Ogedegbe, G., Chaplin, W., Schoenthaler, A., Statman, D., Berger, D., Richardson, T., . . . Allegrante, J. P. (2008). A practice‐based trial of motivational interviewing and adherence in hypertensive African  Americans.  American  Journal  of  Hypertension,  21(10),  1137‐1143. doi:http://dx.doi.org/10.1038/ajh.2008.240 

Page,  S.  J., & Persch, A.  C.  (2013).  Recruitment,  retention,  and  blinding  in  clinical  trials. American Journal of Occupational Therapy, 67(2), 154‐161.  

Polit, & Beck, C. T. (2004). Nursing Research: Principles and Methods: Lippincott Williams & Wilkins. Polit, & Beck, C. T. (2014). Essentials of nursing research : appraising evidence for nursing practice (8th 

ed. ed.). Philadelphia: Lippincott Williams & Wilkins. Polit., D. F. (1996). Data analysis & statistics for nursing research: Appleton & Lange. Riegel, B., Masterson Creber, R., Hill, J., Chittams, J., & Hoke, L. (2016). Effectiveness of Motivational 

Interviewing  in  Decreasing  Hospital  Readmission  in  Adults  With  Heart  Failure  and Multimorbidity. Clinical Nursing Research. doi:10.1177/1054773815623252 

Ruppar, T. M., Cooper, P. S., Mehr, D. R., Delgado, J. M., & Dunbar‐Jacob, J. M. (2016). Medication Adherence Interventions Improve Heart Failure Mortality and Readmission Rates: Systematic Review and Meta‐Analysis  of  Controlled Trials.  Journal  of  the American Heart Association: Cardiovascular and Cerebrovascular Disease, 5(6), e002606. doi:10.1161/JAHA.115.002606 

Sabaté, E. (2003). Adherence to long‐term therapies: evidence for action. Geneva, Switzerland: World Health Organisation. 

Santo,  K.,  Kirkendall,  S.,  Laba,  T.‐L.,  Thakkar,  J., Webster,  R.,  Chalmers,  J.,  .  .  .  Redfern,  J.  (2016). Interventions  to  improve medication adherence  in coronary disease patients: A systematic review  and meta‐analysis  of  randomised  controlled  trials.  European  Journal  of  Preventive Cardiology, 23(10), 1065‐1076. doi:10.1177/2047487316638501 

Shah, R., Desai, S., Gajjar, B., & Shah, A. (2013). Factors responsible for noncompliance to drug therapy in the elderly and the impact of patient education on improving compliance. Drugs & Therapy Perspectives, 29(11), 360‐366. doi:10.1007/s40267‐013‐0075‐3 

Shalansky, S. J., Levy, A. R., & Ignaszewski, A. P. (2004). Self‐Reported Morisky Score for Identifying Nonadherence with  Cardiovascular Medications. Annals  of  Pharmacotherapy,  38(9),  1363‐1368. doi:10.1345/aph.1E071 

Shepherd, C. W., & While, A. E. (2012). Cardiac rehabilitation and quality of life: A systematic review. International  Journal  of  Nursing  Studies,  49(6),  755‐771. doi:http://dx.doi.org/10.1016/j.ijnurstu.2011.11.019 

Short, C. E., James, E. L., & Plotnikoff, R. C. (2013). How Social Cognitive Theory can help oncology‐based health professionals promote physical activity among breast cancer survivors. European Journal  of  Oncology  Nursing,  17(4),  482‐489. doi:http://dx.doi.org/10.1016/j.ejon.2012.10.009 

Sim, J. (1998). Collecting and analysing qualitative data: issues raised by the focus group. Journal of Advanced Nursing, 28(2), 345‐352. doi:10.1046/j.1365‐2648.1998.00692.x 

Smith,  S.  R.,  Rublein,  J.  C.,  Marcus,  C.,  Brock,  T.  P.,  &  Chesney, M.  A.  (2003).  A medication  self‐management program to improve adherence to HIV therapy regimens. Patient Education & Counseling, 50(2), 187‐199.  

Solomon, D. H., Gleeson, T.,  Iversen, M., Avorn, J., Brookhart, M. A., Lii, J.,  .  .  . Katz, J. N. (2009). A blinded randomized controlled trial of motivational interviewing to improve adherence with osteoporosis medications: design of the OPTIMA trial. Osteoporosis International, 21(1), 137‐144. doi:10.1007/s00198‐009‐0951‐9 

Spoelstra, S. L., Given, C. W., Sikorskii, A., Coursaris, C. K., Majumder, A., DeKoekkoek, T., . . . Given, B. A. (2015). A randomized controlled trial of the feasibility and preliminary efficacy of a texting 

  

24  

intervention  on medication  adherence  in  adults  prescribed  oral  anti‐cancer  agents:  study protocol. Journal of Advanced Nursing, 71(12), 2965‐2976. doi:10.1111/jan.12714 

Wald, D. S., Bestwick,  J. P., Raiman, L., Brendell, R., & Wald, N. J.  (2014). Randomised Trial of Text Messaging on Adherence to Cardiovascular Preventive Treatment (INTERACT Trial). PLoS ONE, 9(12), e114268. doi:10.1371/journal.pone.0114268 

Whiting, L. S. (2008). Semi‐structured interviews: guidance for novice researchers. Nursing Standard, 22(23), 35‐40.  

Whittle, J., Yamal, J.‐M., Williamson, J. D., Ford, C. E., Probstfield, J. L., Beard, B. L., .  .  . Davis, B. R. (2016).  Clinical  and  demographic  correlates  of  medication  and  visit  adherence  in  a  large randomized controlled trial. BMC Health Services Research, 16(1), 1‐11. doi:10.1186/s12913‐016‐1471‐x 

Woodruffe, S., Neubeck, L., Clark, R. A., Gray, K., Ferry, C., Finan, J., . . . Briffa, T. G. (2014). Australian Cardiovascular  Health  and  Rehabilitation  Association  (ACRA)  Core  Components  of Cardiovascular Disease Secondary Prevention and Cardiac Rehabilitation 2014. Heart, Lung and Circulation, 24(5), 430‐441. doi:10.1016/j.hlc.2014.12.008 

World Health Organisation. (2003). Adherence to Long‐Term Therapies: Evidence for Action.  Yiend, J. (2010). The effects of emotion on attention: A review of attentional processing of emotional 

information. Cognition and Emotion, 24(1), 3‐47. doi:10.1080/02699930903205698 Zallman, L., Bearse, A., West, C., Bor, D., & McCormick, D. (2016). Patient preferences and access to 

text messaging for health care reminders in a safety‐net setting. Informatics for Health and Social Care, 1‐11. doi:10.3109/17538157.2015.1113177 

Zhu, K.‐F., Wang, Y.‐M., Zhu, J.‐Z., Zhou, Q.‐Y., & Wang, N.‐F. (2015). National prevalence of coronary heart  disease  and  its  relationship  with  human  development  index:  A  systematic  review. European Journal of Preventive Cardiology. doi:10.1177/2047487315587402